Predictive Modelling for Startup Failure

This repository contains Jupyter Notebooks and datasets for a predictive modeling project focused on predicting startup failures. The goal is to compare different machine learning models based on different features, determining the most suitable model to use for deployment and predictive analysis.

Table of Contents

• Project Description
• Files
• Usage

Project Description

Startup failures can be attributed to various factors, and predicting these failures in advance can be crucial for investors, stakeholders, and entrepreneurs. This project leverages machine learning techniques to create predictive models using KNN, Logistic Regression, Random Forest, XGBoost, and AdaBoost algorithms.

Files

• KNN.ipynb: Jupyter Notebook for K-Nearest Neighbors model.
• LOGISTIC REGRESSION.ipynb: Jupyter Notebook for Logistic Regression model.
• RANDOM FOREST.ipynb: Jupyter Notebook for Random Forest model.
• XGBOOST.ipynb: Jupyter Notebook for XGBoost model.
• adaboost.ipynb: Jupyter Notebook for AdaBoost model.
• modelling_df.csv: Processed dataset used for modeling.
• pre modelling.ipynb: Jupyter Notebook for data preprocessing.
• startup_data.csv: Original dataset containing startup information.

Usage

1. Clone the repository: git clone https://github.com/Aynke/MSc-Project.git

2. Install the required dependencies.

3. Navigate to the specific model's notebook you are interested in (e.g., KNN.ipynb) and run each cell to execute the code.

4. Follow the same steps for other models if needed.